Toward Sensor Fault Detection for Autonomous Underwater Vehicles: A Zonotopic Approach

In this work, we consider the detection of velocity sensor faults for autonomous underwater vehicles in a bounded error context. The nonlinearity of autonomous underwater vehicles is handled via a Takagi-Sugeno fuzzy technique. The fault detection is achieved by solving a zonotopic enclosure problem based on zonotopic reachability analysis. To improve fault detection performance, a novel observer structure and H- performance index are introduced. The performance of the presented approach is analyzed utilizing a hidden fault set. Experimental examples are given to validate the effectiveness of the presented approach.